Annual water yield model intersection error?

Dear Nat Cap Team,
I am working with the Invest 3.7.0 water yield model.
Im receiving an error during the run of a Water Yield model.
Attached you can find the log file.
Thank you in advance for the help!
InVEST-Hydropower-Water-Yield-log-2020-03-02–11_59_26.txt (4.5 KB)

Hi @lucy,

Thanks for including the log file in your post. It looks like some of the inputs don’t overlap or intersect in GIS space. Could you check that the eto, precipitation, depth_to_root_rest_layer_path, pawc, lulc rasters do indeed have a common intersection. We usually like to use QGIS or ArcGIS and load our raster files in to see how they compare spatially.

Let us know how it goes and if we can help further!

Doug

1 Like

Hi @lucy -

Also check the coordinate system that is set for each input layer. All input layers must have the same projected coordinate system, and it looks like at least one of them is different (if you look at the “Base list” of coordinates shown in the log file, you’ll see the different values.) I’m not sure about QGIS, but ArcGIS can be confusing, since even if layers are in different coordinate systems, it will try to make them overlap spatially anyway, so it’s not obvious when you just look at the layers. Try reprojecting/Warping the layers with different coordinate systems and see if that helps.

~ Stacie

2 Likes

Hi Doug,

Thanks for your quick response!
Yes,We usually use ArcGIS to finish the drawing.
I will check the coordinates and projection formats of the relevant data and consult you in time if there are any questions.Thank you very much!

Cheers,
Lucy

Hi Stacte,

Thanks for your quick response!
The problem you mentioned is very helpful to me, I will sort out the problems you mentioned one by one.
If I have any questions, I will communicate with you as soon as possible, thank you very much for your guidance.

Cheers,
Lucy

Dear swolny,
I am using the Invest 3.7.0 Scenario Generator model.
Im receiving an error during the run of a Scenario model.
Attached you can find the log file.
Also,only one running successful ,but output do not boundary figure.
Thank you in advance for the help


InVEST-Scenario-Generator-Proximity-Based-log-2020-05-07–09_54_54.txt (1.5 KB)
InVEST-Scenario-Generator-Proximity-Based-log-2020-05-07–09_25_18.txt (4.9 KB)
nearest_to_edge_1.csv (44 Bytes)
nearest_to_edge_11.tif (2.6 KB)

Hi @lucy,

Do you think you could make a new post for this Scenario Generator issue? We try to keep the topics specific to the original problem so that it is easier to search later and easier for future readers to follow. If it’s not too much trouble you could post this identically to a new topic titled: “Scenario Generator ValueError”.

Best,

Doug

I see you already did make a new post! Thanks!

Thank you !
How can I solve this problem?

Hi !

Thank you for having this forum it is an amazing resource :slight_smile:

I am trying to run the Water Yield Model and I beleive I am having an intersection error, or other since the results table show NULL values for AET, and wyield _mm values (image below)

I have attached a folder with the area outline, the result from the root depth and plant water , biophysical table and the log.Help.zip (397.1 KB)

  • Do you think the intersection error its due to a resolution issue?

  • Or am i doing a mistake with the data sources or extraction?

  • Error in the biophysical table?

Thank you so much for your help!

Laura

I am using the following datasources:

Precipitation
From Fick, S.E. and R.J. Hijmans, 2017. WorldClim 2: new 1km spatial resolution climate surfaces for global land areas. International Journal of Climatology 37 (12): 4302-4315.

Evapotranspiration
From Trabucco, Antonio; Zomer, Robert (2019): Global Aridity Index and Potential Evapotranspiration (ET0) Climate Database v2. figshare.

Depth root
From : FAO/IIASA/ISRIC/ISSCAS/JRC, 2012. Harmonized World Soil Database (version 1.2). FAO, Rome, Italy and IIASA, Laxenburg, Austria.

Plant water available fraction
From
Batjes NH 2016. Harmonised soil property values for broad-scale modelling (WISE30sec) with estimates of global soil carbon stocks. Geoderma 2016(269), 61-68(doi: 10.1016/j.geoderma.2016.01.034),with supplemental informatio

At: https://data.isric.org/geonetwork/srv/eng/catalog.search#/metadata/3a9ed87d-affc-4f72-aa6e-72db4fefec40

KC: from INVEST kc calculator

The reclass txt from MOOC on Ecosystem
Services : a Method for Sustainable Development University
of Geneva (UNIGE)

Hi @lauzura -

Thanks for sending all of this information. Since some of the sub-watersheds do have values for AET and wyield, look at what’s different between the ones that do have results and those that don’t. Is it possible that one (or more) of your input rasters don’t have values in those watersheds? If one layer doesn’t have values, you will get NoData in the result, since the model cannot do the calculations properly.

Looking at the data you sent in ArcGIS, and it appears that SuelosGarabitoWHSCRTM05reclass_depth2root.tif.tif has values of mostly NoData within the border of CantonGarabitoBordersinrios1.shp. This would certainly cause the model problems, and produce NoData in the result.

It would have been useful to see all of your inputs, particularly the sub-watersheds that were used as inputs, so I can’t debug further. But I do advise revisiting the root depth layer and see what happens when your area of interest is entirely covered by valid floating point values instead of NoData. And check your other inputs to make sure they have full coverage across your watersheds too.

~ Stacie

1 Like

Hi Stacy!
Thank you so much for your response!

I have attached the watershed file.
You are right there’s no data in the whole raster… When I see the world data available ( in the links below ) i see values for the study area… but when i cut the layer and do the reclass I lose them…

do you think is a precision issue?
or am I reclassing incorrectly?
would you recommend to use other data source?
do you know one where i can have that info at that scale?

Thak you so much for your support,

Laura

Help2 (2).zip (70.2 KB)

Depth root:
From : FAO/IIASA/ISRIC/ISSCAS/JRC, 2012. Harmonized World Soil Database (version 1.2). FAO, Rome, Italy and IIASA, Laxenburg, Austria. At: https://webarchive.iiasa.ac.at/Research/LUC/External-World-soil-database/HTML/HWSD_Data.html?sb=4

Plant water available fraction
From
Batjes NH 2016. Harmonised soil property values for broad-scale modelling (WISE30sec) with estimates of global soil carbon stocks. Geoderma 2016(269), 61-68(doi: 10.1016/j.geoderma.2016.01.034),with supplemental informatio

At: https://data.isric.org/geonetwork/srv/eng/catalog.search#/metadata/3a9ed87d-affc-4f72-aa6e-72db4fefec40

Hi @lauzura -

What are you doing to “reclass” the soil data? If there is soil data in your study area in the original global layer, then it should appear after you clip it to the study area. Remember that since the layer is coarse, and we need it to cover the whole watershed, it’s best to clip the data using a watershed that has a buffer around it, to make sure that the watershed is completely covered by input data. Having extra data outside of the edges of the watershed is fine, but we don’t want to have any missing data (NoData) cells within the watershed.

The HWSD is pretty coarse. For global data, we usually use ISRIC Soil Grids.

~ Stacie

Hi Stacie!

Thank you for your help!
To reclass i am using the r.reclass tool with the .txt i have attached to the first post folder (one for root and other for plant water)

I am going to try the ISRIC Soil Grids thank you, does it need any other preprocessing for plant water availability and root depth? (Soils its not my strenght at all)

Thank you again,

Laura

Hi @lauzura -

I haven’t had to process SoilGrids in a while, and looking at the latest version (2.0) it appears that some of the properties that were provided in the previous version aren’t there yet, including soil depth and available water content. But you can try the previous version (1.7) - the layers are here and a FAQ that describes each layer is here.

For AWC, it looks like their units are in percent, so you’d just need to convert percent to a fraction to have the units required by InVEST. The trick is that they provide the AWC layer at multiple depths, so you need to process them to get a single value, which I haven’t needed to do. The FAQ does provide a link to this paper, which gives an equation that can be used for general averaging across horizons.

@jandrewjohnson, I think you went through the process of processing Soil Grids to a single PAWC map - can you give us any tips? This topic has come up before, so it would be really useful to have this info as a searchable resource here - Thanks!

~ Stacie

1 Like